PrimeGrid
Please visit donation page to help the project cover running costs for this month

Toggle Menu

Join PrimeGrid

Returning Participants

Community

Leader Boards

Results

Other

drummers-lowrise
1) Message boards : Problems and Help : CPU usage and Linux (Message 118665)
Posted 834 days ago by Kwartet!Project donor
Well, five threads in a twelve-thread system is 41.7%. Five Primegrid processes at 8% is 40%. I don't see much discrepancy.

My Ubuntu system says [process monitor] 48% usage for SR5 running 4 threads, [resource monitor] sees eight threads active at 50% on average. Resource monitor is really dynamic, switching threads from core to core, but I believe that is what it is trying to tell me.
2) Message boards : Sieving : Recerve 2nd GPU for GFN Sieving, means exclude 2nd GPU on BOINC (Message 116836)
Posted 905 days ago by Kwartet!Project donor
Hi klepel,

In the manual dukebg pointed to, I can find no other option than have BOINC stopping *all* GPU work, if some important program is running.

I think the only way is to run two instances of BOINC concurrently. Then, let one BOINC instance use GPU 0, the other GPU 1. Now, stopping GPU work, as you already know how to do, should only affect one of the GPUs.

I do personally not have any experience with this stuff, but it can be done. On the net, I found this how-to: http://www.overclock.net/forum/365-overclock-net-boinc-team/1628924-guide-setting-up-multiple-boinc-instances.html

Best wishes
3) Message boards : Project Staging Area : 2017 PRPNet September Challenge (Message 110525)
Posted 1093 days ago by Kwartet!Project donor
Ich denke, Grebuloner meint: http://u-g-f.de/PRPNet/formerly.php
4) Message boards : Project Staging Area : Unlikely runtime on Core2Duo E8500 (Message 110442)
Posted 1097 days ago by Kwartet!Project donor
Ok, thanks. It did surprise me my oldish E8500 could be so fast, relatively speaking.
5) Message boards : Project Staging Area : Unlikely runtime on Core2Duo E8500 (Message 110391)
Posted 1099 days ago by Kwartet!Project donor
So, I switched fom 27 to FPS. Client log:

[2017-09-22 06:39:17 W(] 27: Getting work from server prpnet.primegrid.com at port 12006
[2017-09-22 06:39:18 W(] 27: PRPNet server is version 5.4.0
[2017-09-22 20:13:49 W(] 27: 27*2^6761330+1 is not prime. Residue C2BE76B41CB710A5
[2017-09-22 20:13:49 W(] Total Time: 34:19:26 Total Work Units: 3 Special Results Found: 0
[2017-09-22 20:13:50 W(] 27: Returning work to server prpnet.primegrid.com at port 12006
[2017-09-22 20:13:51 W(] 27: INFO: Test for 27*2^6761330+1 was accepted

[2017-09-22 20:13:51 W(] 27: INFO: All 1 test results were accepted
[2017-09-22 20:13:51 W(] Due to changes in the prpclient.ini file, work classes have been reloaded
[2017-09-23 01:20:51 W(] FPS: 243931!+1 is not prime. Residue FE20AD443C06BC26
[2017-09-23 01:20:51 W(] Total Time: 39:26:28 Total Work Units: 4 Special Results Found: 0
[2017-09-23 01:20:52 W(] FPS: Returning work to server prpnet.primegrid.com at port 12002
[2017-09-23 01:20:53 W(] FPS: INFO: Test for 243931!+1 was accepted
[2017-09-23 01:20:53 W(] FPS: INFO: All 1 test results were accepted
[2017-09-23 01:20:53 W(] FPS: Getting work from server prpnet.primegrid.com at port 12002
[2017-09-23 01:20:55 W(] FPS: PRPNet server is version 5.4.0

5 hours between reloading classes and returning 243931!+1. While a 27 test takes some 14 hours. Also note: I seem not to have gotten work from port 12002 for that test. Must be something not OK?
6) Message boards : Number crunching : TRP-Sieve HAS ENDED!!! (Message 105414)
Posted 1307 days ago by Kwartet!Project donor
Looks like a test run; 185 errors on a machine with 272 cores is not that bad.
7) Message boards : Problems and Help : Suspending computaion - no recent user activity (Message 105288)
Posted 1311 days ago by Kwartet!Project donor
Something changed in your power management settings?
8) Message boards : Problems and Help : Immediate Computation Error on OCLcudaGFN% (Message 104626)
Posted 1330 days ago by Kwartet!Project donor
Xialeth, you nailed the problem: "I don't have OpenCL installed"

You may check what's installed for your videocard with GPU-Z, to be downloaded from here: https://www.techpowerup.com/download/gpu-z/

Easiest solution is to follow Michael's advice and reinstall the NVidia driver from here: http://www.nvidia.com/download/driverResults.aspx/114351/en-us

Maybe there is an option to download and install just yhe OCL part, I don't know.
9) Message boards : Problems and Help : Not reliably receiving tasks (Message 102923)
Posted 1364 days ago by Kwartet!Project donor
I noticed the same behaviour, hence maybe the same problem on my system.

BOINC scheduler seems to give processor time to each project *in turn*, for an amount, or equivalent, set in BOINC Manager (x% of processor time).

If your project cannot fill *its turn* with enough WUs, too bad. Scheduler will not use the void for WUs from other (sub)projects, even if they are waiting on your pc.

This may happen, if you have limited the number of concurrent WUs. Solution: relax the constraint on concurrency somewhat.

Edit: BOINC version 7.6.33
10) Message boards : Problems and Help : Project Memory Requirement (Message 102749)
Posted 1369 days ago by Kwartet!Project donor
And Hi to you.

Wuprop collects this type of data, see for example

[url]http://wuprop.boinc-af.org/results/projet.py?projet=PrimeGrid&application=321+%28LLR%29
[/url]

In order to view other (sub)projects, mouseover on (?) afficher le menu.

Best regards,


Next 10 posts
[Return to PrimeGrid main page]
DNS Powered by DNSEXIT.COM
Copyright © 2005 - 2020 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 3.52, 4.12, 3.79
Generated 26 Sep 2020 | 18:36:39 UTC